Should You Parlay Favorites or Underdogs?

Both styles of parlay are fundamentally bets against compounded vig. Whether stacking favorites or underdogs is "better" depends entirely on whether the individual legs carry positive expected value. Most do not, which is why most parlays lose.

Here is the math, the trap on each side, and the narrow conditions where parlays actually make sense.

The structural problem with all parlays

Every leg of a parlay must hit for the parlay to win. The hit probability is the product of the individual probabilities, but the payout is calculated by multiplying the decimal odds. The vig in each leg multiplies along with everything else.

A 2-leg parlay of two -110 lines pays roughly +264 (combined decimal odds 3.64). True 50/50 on each leg gives 25% combined hit rate. Breakeven at +264 is 27.47%. Even a 50% true win rate on each leg loses money on the parlay.

For a parlay to be +EV, the true probability of all legs hitting must exceed the breakeven implied by the offered price. With each added leg, the gap widens.

The all-favorite parlay

The most common bad parlay. Three or four heavy favorites stacked together. Looks safe. Pays poorly. Loses on the math.

Consider three favorites at -200 each. Each implies a 66.67% win rate. Combined hit probability is 0.667 ^ 3 = 29.7%. The combined American odds at standard pricing are roughly +237. Breakeven at +237 is 29.67%.

The bet is essentially zero EV before you account for any of the favorites being soft. If even one of those favorites is actually 64% true probability instead of 67%, the parlay flips to negative EV. The cushion is paper-thin.

People love these parlays because the hit rate feels safe. Three out of four times, the parlay cashes for a small return. The fourth time wipes out three previous wins. Net result over a long sample: roughly flat or slightly down.

The all-underdog parlay

The other extreme. Stacking three +200 dogs to chase a +2700 payout. Looks juicy. Almost never hits.

Three +200 underdogs each imply a 33.33% win rate. Combined hit probability is 0.333 ^ 3 = 3.7%. Combined American odds at standard pricing are roughly +2700. Breakeven at +2700 is 3.57%.

Again, essentially zero EV. The wins feel enormous when they happen, which creates the illusion of a winning strategy. They happen so rarely that the long-run return is similar to the all-favorite parlay: a small grind toward zero.

When 2-leg beats 4-leg

The cleanest version of "small parlay good, large parlay bad" comes from how vig compounds. Each leg you add multiplies your hit rate down by the win probability and multiplies the payout up by the decimal odds. Both effects are present, but the vig grows non-linearly.

A 2-leg parlay of two +EV bets retains most of the underlying edge because there are only two compounded vigs to overcome. A 4-leg parlay of the same +EV legs adds two more vigs. Unless every single leg carries significant edge (say, 5%+ EV each), the additional vig drags the combined bet into negative territory.

Practical rule: if you cannot articulate why each leg is +EV as a straight bet, do not include it in a parlay. The parlay does not rescue weak legs. It punishes them.

Correlation: the only structural edge

Sportsbooks price parlays as if the legs are independent. Sometimes they are not. When two outcomes are positively correlated, the true joint probability is higher than the product of the individual probabilities. That gap is your edge.

Classic positive correlations:

• Team moneyline + team total over (winning teams tend to score)
• QB passing yards over + team total over (a high-scoring game inflates both)
• Game total over + first half total over (high-scoring games are high-scoring early)

Books block the most obvious correlations through "no same game parlay" rules or special pricing on same-game parlays. The remaining residual correlation is small but real. Identifying it is the only way to consistently profit from parlays.

Negative correlations work in the book's favor and should be avoided. Stacking the over with both team unders is mathematically self-defeating.

When parlays make sense

Three narrow cases:

1. Two or three +EV straight bets you would take anyway. Parlaying them concentrates the action when bankroll size limits your straight wagers. The expected return drops slightly versus separate straights, but the variance also concentrates. This is a tradeoff, not an edge.

2. Genuine correlation the book has not priced in. Same-game parlays where the legs share a positive driver and the price does not fully reflect it. Rare, but real.

3. Bankroll preservation under a maximum bet limit. If your local book caps you at $100 per straight bet but allows $100 parlays, parlaying lets you put more total dollars at risk on plays you like. This is a logistical workaround, not a strategic edge.

When parlays never make sense

• As a way to "spice up" the slate when you have no reads
• As a way to chase losses with a high-payout dart
• As a way to bet legs you would not bet straight
• As anything containing a leg you have not modeled

Sportsbooks heavily promote parlays for a reason. Parlay hold percentages run two to four times higher than straight bets. The book is happy to take action on +2700 dreams all day.

The disciplined approach

If you are going to play parlays, keep them small (2-3 legs maximum), insist on positive EV on every individual leg, and look actively for correlation in your favor. Track parlay performance separately from straights. Most bettors who do this honestly find parlays drag their overall ROI down.

The straightforward alternative is to bet each +EV leg as a straight. The expected return is higher and the variance is lower. The only reason not to is if you genuinely believe you have identified correlation. In which case, size accordingly and verify with results over time.

The daily card caps parlays at three legs and only ships them when the underlying legs are independently +EV. The track record breaks straights and parlays apart so you can see the actual hit rates.

Frequently asked questions

Are parlays of favorites or underdogs more profitable?

Neither is automatically profitable. Both face the same compounded vig. Favorite parlays hit more often but pay much less; underdog parlays pay more but hit far less. The math typically favors smaller parlays (2-3 legs) where individual bets carry positive EV.

Why do all-favorite parlays underperform?

Compounded juice eats the small payouts. Three -200 favorites at roughly 67% each multiply to 30% combined hit rate, paying about +237. The break-even hit rate is 30%, so there is essentially no edge unless you have substantial probability advantage on each leg.

Are 2-leg parlays better than 4-leg parlays?

Almost always, if your goal is positive expected value. Each added leg multiplies your hit rate down and the vig compounds. A 2-leg parlay of two +EV bets keeps the edge largely intact. A 4-leg parlay needs every leg to carry significant edge to overcome the math.

What is correlation in parlays and why does it matter?

Two outcomes are correlated when one happening changes the probability of the other. A team moneyline plus team total over is positively correlated; both depend on that team scoring. Sportsbooks usually block obvious correlations, but residual correlation in your favor is the only structural edge inside a parlay.

When should you avoid parlays entirely?

When you cannot identify positive EV on each individual leg, when the parlay includes legs you would not bet straight, or when you are using parlays to chase a bad day. Parlays are not a way to manufacture edge from -EV bets; they amplify whatever edge or leak is in the underlying legs.